[Gluster-users] Please Help- XenSever - "The NFS server version is unsupported Check your settings and try again"

[Matthew Mackes]

Hi Dan,

So I took your advice, and I have added all of the machines- 2 Gluster
Storage Machines, and 6 Xen Host Machines to each machines hosts file.
Now all machines are pingable by name.

I am afraid that did not correct the issue.

I have a more basic question. Should both machines be reachable via
NFS if they are set to replicate between the two.

Because I'm not sure this is as much a Xen issue, more over I am
unable to connect to the second machine once I have changed the port
number (to meet the Xen requirement)

Thoughts?

Matt


On Mon, Feb 13, 2012 at 1:17 PM, Daniel Kradolfer <ml at smilesolutions.ch> wrote:
> Hi Matt,
>
> gluster is listening on port 2049. that's ok. the storage servers has to be
> able to lookup hostnames of the IPs connecting to gluster nfs server. if
> gluster-node isn't able to resolve xen-servers hostname, connection will
> fail.
>
> Dani
>
>
> Am 13.02.2012 17:13, schrieb Matthew Mackes:
>
> Hi Dan,
>
> Thanks for your reply.
>
> I am able to ping and ssh the server from all Xen Host machines in the
> Xen pool. It seems to be up, and reachable at the OS level.
>
> Here is the output from lsof:
> glusterfs 1466  root    8u  IPv4  79120      0t0  TCP *:2049 (LISTEN)
>
> I think that Gluster is responding on this machine because Xen is
> reporting " The NFS server version is
> unsupported." I think that Xen would report something else if it could
> not contact the machine at all.
>
> Thanks Dan, please keep brainstorming! I appreciate it!
>
> Matt
>
>
>
> On Mon, Feb 13, 2012 at 11:01 AM, Daniel Kradolfer <ml at smilesolutions.ch>
> wrote:
>
> Hello Matt
>
> the storage server has to be able to resolv the ip-address of the
> xen-server. this may be a problem. i also had problems to get gluster really
> listening on port 2049. if 'lsof -i -n -P | grep 2049' doesn't show gluster
> listening on this port, I would reboot the whole cluster.
>
> Freundliche Grüsse
>
> Daniel Kradolfer
>
>
> Am 13.02.2012 16:48, schrieb Matthew Mackes:
>
> Hello Everyone,
>
> I am having an issue that I hope someone might have some information about.
>
> I have a Xen Server Version 6 Host Cluster that I am attempting to use
> Gluster as the NFS storage repository for.
>
> I have two Gluster storage servers, both are running the latest
> stable, and CentOS 6.2.
>
> Both Gluster storage servers are setup for Replicated Volumes, and are
> working well.
>
> In order for XenServer to allow me to mount the Gluster NFS, I have
> set using one of the Gluster Machines (The first) to port 2049. This
> change instantly replicated to the second Gluster server in the pool.
>
> (First server (Working with Xen))
> gluster> volume info
>
> Volume Name: xen-volume
> Type: Replicate
> Status: Started
> Number of Bricks: 2
> Transport-type: tcp
> Bricks:
> Brick1: 10.10.135.10:/BRICK-STORAGE
> Brick2: 10.10.135.11:/BRICK-STORAGE
> Options Reconfigured:
> nfs.port: 2049
> gluster>
>
> (Second server (Not Working with Xen-"The NFS server version is
> unsupported Check your settings and try again" )
>
> gluster> volume info
>
> Volume Name: xen-volume
> Type: Replicate
> Status: Started
> Number of Bricks: 2
> Transport-type: tcp
> Bricks:
> Brick1: 10.10.135.10:/BRICK-STORAGE
> Brick2: 10.10.135.11:/BRICK-STORAGE
> Options Reconfigured:
> nfs.port: 2049
>
>
> Here is the issue.
>
> If I attempt to mount the first Gluster storage server in the Xen
> Administrative Console (Xen center), the Scan button reports an NFS
> error, but if I choose finish, it does mount the storage, and it works
> very well. The vm's also replicate to the second gluster storage
> server.
>
>  If I attempt to mount the second Gluster storage server in the Xen
> Administrative Console (Xen center), I receive the error back:
>
> "The NFS server version is unsupported Check your settings and try again.:
>
> I receive this error for the second server, even when:
> 1. The first server was never mounted
> 2. Both servers are a clean new installs, and contain no data.
> 3.Both installs are CentOS base installs, and I add in packages as I need
> them
>
> So, the first server always mounts (but with an error during the test
> scan), the second never does, giving the error NFS version is
> unsupported.
>
> *The second server also cannot be mounted as a ISO NFS share.
> * The second server also cannot be mounted on a normal Fedora 16
> workstation (My workstation).
> It seems as if once the NFS port has been changed in Gluster the
> second server (in replication) becomes inaccible via NFS)
>
>
> I hope someone has some ideas. Thank you everyone for reading and your help.
>
> Matt
>
>
>
>
> _______________________________________________
> Gluster-users mailing list
> Gluster-users at gluster.org
> http://gluster.org/cgi-bin/mailman/listinfo/gluster-users
>



-- 
Matthew Mackes
Delta Sonic CWS, Buffalo N.Y.
716-541-2190

---------------------------------------------------------------
The Steps to solve any problem (Scientific Method):

1. Ask a Question
2 .Do Background Research
3. Construct a Hypothesis
4. Test Your Hypothesis by Doing an Experiment
5. Analyze Your Data and Draw a Conclusion
6. Communicate Your Results
---------------------------------------------------------------